The Impact of Adjuvant Chemotherapy in Pulmonary Large Cell Neuroendocrine Carcinoma (LCNC)

Disclosure None

Background Torino, Italy

LCNC Rare tumor (2% to 3% all resected primary lung cancers) Preoperative diagnosis is often impossible Clinical behavior and prognosis similar to SCLC Surgery alone is insufficient to treat LCNC, even in early Stages

Biologically related to Neuroendocrine tumors, but still considered a variant of LCCs, and therefore, accordingly treated

Author Year N pts 5-y OS 5-y OS Stage I Recurrence % Garcia-Yuste et al. 2000 22 21 % 33 % 59 % Takey et al. 2003 87 57 % 67 % 40 % Battafarano et al. 2004 45 30 % 33 % 49 % Paci et al. 2004 48 21 % 27 % NA Rossi et al. 2005 83 NA 33 % 65 % Veronesi et al. 2006 144 42 % 52 % 40 % Sarkaria et al. 2011 100 NA 58 % 38 %

LCNC Due to its rarity and the lack of RCT, the optimal treatment is still debated Few clinical papers evaluated the role of induction/adjuvant CT

3 years ago The ESTS Lung NETs-WG was created with the aim to: create a group of physicians worldwide expert in NETs develop scientific knowledge on such rare neoplasms

2014: 14 Centres 2054 patients 1% 12% 16% Atypical Carcinoid LCNC Mixed Tumor SCLC Typical Carcinoid 61% 5% 5% Unspecified

Aim of the study Torino, Italy

Aim of the study To evaluate the possible adjuvant CT effect on LCNC survival To assess clinicopathologic prognostic factors in a surgicallybased population of patient with LCNC

Methods Torino, Italy

460 patients with LCNC surgically treated between 1992-2014 in 14 Centers 28 patients with missing information on vital status Study Flow-Chart 432 Patients 32 patients with missing information concerning adjuvant chemotherapy 400 patients median FU: 38 months % FU completeness: 94%

LCNC diagnosis All the histological samples were reviewed by local NETs expert Pathologists LCNC definitive diagnosis was made according to: 2004 WHO Lung Tumors Classification criteria Travis histological guidelines for NETs diagnosis

NET morphological/immunohistochemical characteristics: Synaptophysisn Chromogranin A Neuroendocrine morphology (organoid nesting, palisading rosettes and trabeculae High mitotic rate (11 or more) Abundant necrosis Large cell size Low nuclear/citoplasm ratio Tumor cells positive for neuroendocrine markers: Synaptophysisn Chromogranin A CD56

Demographics Age Gender Smoking habit Clinical variables Previous malignancy Tumor site PS (ECOG) Tumor-related variables Stage (TNM 7 edition) Treatment variables Type of surgical resection Completness of resection Use of chemotherapy Retrospective multicentre study Outcome measure : overall survival (OS) Cox proportion hazard model with shared frailty (for center heterogeneity)

Results Torino, Italy

400 pts Age (median IQR-) 66 (58-72) No. % Gender (male) 252 63 Smokers (current/former) 99 25 Previous malignancy 99 25 ptnm I 185 48 II 110 29 III 76 20 IV 12 3 Induction therapy 53 (44 CT; 9 RT) 13 Adjuvant CT 146 37 Adjuvant RT 38 10

Type of surgery Mediastinoscopy Extended Resection Pneumonectomy Bilobectomy Sleeve lobectomy 1 14 9 7 43 Lobectomy 263 Segmental resection Wedge Resection R0 resection: 360 cases (97%) 24 38

Median FU: 38 months FU completeness: 94% 213 patients died (69 in the adjuvant CT group)

Overall survival 1.00 0.75 Median OS: 43 months 0.50 0.25 3-y surv rate: 54 % 5-y surv rate: 45 % 0.00 At risk: 0 12 24 36 48 60 72 84 96 108 120 Months 400 292 202 149 110 94 70 54 35 25 22

OS according to adjuvant CT administration 1.00 0.75 A slight improvement in OS was observed in those who received adjuvant CT (HR 0.82; 95%CI: 0.62-1.09, P=0.17) 0.50 0.25 0.00 At risk: No Yes No Yes 0 12 24 36 48 60 72 84 96 108 120 Months 254 182 126 92 68 59 42 33 22 19 17 146 110 76 57 42 35 28 21 13 6 5

Multivariate analysis Variable HR (95%CI) P Adjuvant Chemoterapy YES vs NO (adjusted for the below factors) 0.74 (0.53 to 1.02) 0.06 Age (per 1 year increase) 1.03 (1.01 to 1.04) 0.001 Male gender 1.18 (0.87 to 1.59) 0.28 Previous Malignancy 1.01 (0.73 to 1.39) 0.94 ECOG PS>=2 1.62 (1.21 to 2.16) 0.001 Vascular Invasion 1.29 (0.95 to 1.77) 0.11 ptnm II vs I 1.4 (0.98 to 2.01) 0.06 III/IV vs I 2.57 (1.76 to 3.76) <0.001 Year of Surgery 2000-2007 vs 1992-1999 0.74 (0.49 to 1.12) 0.15 2008-2014 vs 1992-1999 0.61 (0.38 to 0.95) 0.03 An evidence of a higher OS in adjuvant CT was observed

Variable HR (95%CI) P-Interaction Overall 0.74 (0.53 to 1.02) Age 0.64 <65 0.69 (0.43 to 1.1) >=65 0.8 (0.52 to 1.21) Gender 0.6 Female 0.83 (0.49 to 1.42) Male 0.7 (0.48 to 1.02) Previous Malignancy 0.34 No The subgroup analysis did not 0.8 (0.56 show to 1.14) Yes 0.56 (0.29 to 1.08) Vascular Invasion 0.25 No 0.65 (0.4 to 1.06) Yes 0.96 (0.6 to 1.56) ptnm 0.99 I 0.71 (0.36 to 1.41) II 0.75 (0.44 to 1.28) III-IV 0.76 (0.44 to 1.3) Subgroup Analysis for the effect of administration of adjuvant Chemotherapy any subset of patients that significantly benefited from adjuvant CT administration

Conclusions Torino, Italy

We observed a signal of an improved survival in LCNC patients treated with adjuvant CT We did not identify a particular subset of patients in which adjuvant CT might be more appropriate Prospective data collection (ESTS prospective database), will hopefully help to define more tailored treatment strategy for such aggressive neoplasm

Thank you very much for your attention Torino, Italy

Cause of death Unknown 30 Tumor related 130 Treatment complications 5 Non tumor related 48

Local recurrences: 61 (26 in the adjuvant CT group) Distant MTS: 140 (59 in the adjuvant CT group)

Stage I OS 1.00 0.75 I 0.50 0.25 3-y surv rate: 63 % 5-y surv rate: 51 % 0.00 At risk: 0 12 24 36 48 60 72 84 96 108 120 Months 185 146 98 75 52 46 31 25 18 13 12

Statistical analysis Kaplan-Meier method: probabilities of survival Cox proportional hazards regression model with shared frailty: effect of adjuvant CT on OS Effect modifications by subgroups: inclusion in the model an interaction term between the covariate indicating adjuvant CT and the subgroup covariate of interest, adjusting for the others

Study limitations: retrospective and multicenter design long recruitment period possible inherent treatment selection bias (pts receiving CT may have been selected among those with better clinical/functional conditions)

Future directions: Future clinical trials might be "ad hoc"designed and new (biological) drugs might be tested to treat such aggressive neoplasms

LCNC: Due to its rarity and the lack of RCT, LCNC s optimal treatment is still debated Few clinical papers evaluated the role of induction/adjuvant CT

Recent increased incidence: 2.3 2.8 cases/100,000/year (diagnostic techniques improvement; lung cancer screening programs diffusion) Yao JC: J Clin Oncol. 2008 Jun 20;26:3063-72